2020-04-30 03:23 AM
Does anybody know a board with a STM32F4 of F3 with the following features:
A display would be optional.
Had been looking at the usual suspects (ST, Olimex, Waveshare, Sparkfun, Mikroe), but didn't find anything appropriate.
I prefer a MCU with SPL code available, which limits the selection to the F1, F2, F3 and F4 series.
The (non-commercial) application should do burst collection of analog data, and transfer it to PC or store it right away.
2020-04-30 04:02 AM
STM3240-EVAL
STM32F469I-DISCO
2020-04-30 07:11 AM
Yes, thanks.
Albeit the STM3240-EVAL is quite a bit over-equipped and overpriced for my purpose.
The F469-Disco looks o.k. so far - only ST's availability numbers seem to lie recently.
Checked some DISCO boards recently (including similar F7 ones) at the distributor's pages (Farnell, Digikey, Mouser), and none is currently available.
Right now, I can start with the F4-Disco, and port later in a matter of hours.
Tried with a F746 Nucleo, only the LL code is quite a disappointment. Especially in regard to documentation and examples.
The effort is basically the same as with direct register access.
2020-04-30 08:43 AM
2020-04-30 09:43 AM
Nice board at affordable price!
2020-04-30 10:58 AM
Guess some one what a piece of ST's lunch
2020-04-30 11:01 AM
Does it actually need external RAM, internal's faster, what rate are you sampling, couldn't you just spill to the sd card using a 64KB DMA ping-pong arrangement?
TaoBao has had some cheap/compact boards that would work as streaming data recorders.
2020-04-30 02:47 PM
This one looks interesting, doesn't it:
https://ae01.alicdn.com/kf/HTB1LnbgKFXXXXcvXXXXq6xXFXXXF/201991911/HTB1LnbgKFXXXXcvXXXXq6xXFXXXF.jpg
(I'm not endorsing it, no personal experience, just googled)
JW
2020-04-30 03:37 PM
Or this one
https://mikrocontroller.bplaced.net/wordpress/?page_id=568
Pity it's not a commercial project.
There might've been some interest in the 'F4 back then some 5 years ago, it all faded away since then...
JW
2020-05-01 03:32 AM
> Does it actually need external RAM, internal's faster, ...
That's true, and I plan to start out that way.
The device is intended for experimenting with spacial maps, created from acoustic echos. Kind of sonar.
Which means at least 3 channels at rates below 50kHz.
Range is thus directly related memory depth / buffer size.
Never tried writing audio streams directly to SD. For the F4-Disco + BB, the SD interface is basically SPI, not sure if that works out.
The F4-Disco has enough resources and performance to try the concept. And I have several projects to pick up code from, so I can have a basic prototype in one day.
Ethernet and USB would be options for streaming. I lack any experience with USB audio streaming on both sides, so I would go for ethernet first.
Audio capturing is the simpler part.
Noise filtering and data processing is more complex, and resource consuming. I want to start out with a PC first.
For a MCU implementation, external RAM and a display would be essential.